Implementing FHIR Search: Pagination and Total Count Retrieval

Details

This article serves as an implementation guide for the FHIR search capabilities within Google Cloud's Cloud Healthcare API. The API is designed with high scalability and performance while complying with REST and FHIR standards. Key technical considerations addressed are 'pagination' (dividing results) and 'total count retrieval'. 1. Sequential Browsing: Suitable for low-latency applications where users view results one by one. This workflow involves obtaining an approximate total count on each page and using pagination links to retrieve data continuously. 2. Batch Processing: Ideal for acquiring a complete set of search results for subsequent analysis or processing. While the accurate total count is only confirmed on the final page, all pages can be sequentially called to gather the full dataset. 3. Advanced Design (Near/All Results): For advanced UX, the article proposes caching results into 'application pages' and loading surrounding results based on user interaction (similar to Google Search). This maintains low latency while handling large volumes of data. Specific design options are detailed, including controlling page size using the `_count` parameter and limiting fields to retrieve only necessary identifiers.